If you like me have quite frequent problems with underexposure in an rather unpredictable way on your K5 or other Pentax DSLRs(in my case especially with bright aperture lenses like older 1.4 lenses etc) and dont want to keep changing exposure compentation when you changing from big apertures to small and a lot of other scenarios, then I have a suggestion which have helped me a lot:I always shoot RAW and jpg at the same time and often want jpgs to share straight away,for instance print out while at a party or in a wedding,or email at once.Before I used exposure compentsation,and then from time to time blew out the highlights ,Now I go in and choose say Bright or Vivid in the shooting menu, there I choose highkey-lowkey adjustment and say have it turned to give a bit brighter results in jpg. Now I can print jpgs straight out,and with K5s fantastic files,I can get really good quality files from RAWs anyway. While before when using normal exposure compensation,I often blew jpgs and RAW at the same time,and had to save results in some pictures of important moments for instance by making a black and white form the bluechannel etc. That often works even if it seem totally overexposed,sometimes in combination with whitebalance as well.The other way around, if you really know the shooting conditions,and that you are not likely to blow highlights,or can re-shoot if it happens, and prefer to expose the histogram to the right, then you can have another jpg engine setting put on say Portrait and dial in lowkey adjustment to also have perfect OOCjpgs to use straight away. Easy to switch between also.when workin in TAv and and some other modes I also prefer to have auto iso dialed strating not from iso 100 but at least 125 or more,as that gives more dynamic range(or room for error),the way I shoot.Any body who have more experince with that? (Dynamic range at different ISOs)By the way,I think K-01 jpgs make very much sense from this angle of view,no need to be conservative on jpgs if you also shoot RAW,only thing I disagree with is that there is a bit (more) noise removal in RAW high iso,so for high iso combined with very much detail,I still prefer K5.Exposure is more accurate or easy in K-01,and focus more accurate with say FA77mm (and other lenses where I got both front and back focus on K5)at bright apertures,for me anyway.So less work for quick sharing.And mostly overall better results in normal iso,which for me is at least to 1600,and only when very small details in high iso,not as good for me.In low light, K5 have quiker autofocus,and works at lower light levels,and with less contrast.Also critical focus handheld much better on K-01.lowlight macro ir

henning kraggerudwrote:when workin in TAv and and some other modes I also prefer to have auto iso dialed strating not from iso 100 but at least 125 or more,as that gives more dynamic range(or room for error),the way I shoot. Any body who have more experince with that? (Dynamic range at different ISOs)Raw DR always reduces as you increase ISO. Here's the DxOMark chart for the K-5. http://www.dxomark.com/index.php/Cameras/Camera-Sensor-Database/Pentax/K5The reduction ought to be about 1 stop of DR lost for each stop of increased ISO, although this direct relationship doesn't always hold, but DRneverincreases as you increase ISO.In-camera JPG conversion throws away some of the raw DR, reducing it to about 9 stops. As the raw DR drops to about 9 stops at ISO1600, you'll get about 9 stops DR in JPG whatever ISO you set. -- ---GerryFirst camera 1953, first Pentax 1985, first DSLR 2006 http://www.pbase.com/gerrywinterbourne

Gerry Winterbournewrote:henning kraggerudwrote:when workin in TAv and and some other modes I also prefer to have auto iso dialed strating not from iso 100 but at least 125 or more,as that gives more dynamic range(or room for error),the way I shoot. Any body who have more experince with that? (Dynamic range at different ISOs)Raw DR always reduces as you increase ISO. Here's the DxOMark chart for the K-5.Thanks for the reply I assume that is not for extended iso,so from 80 to 100 you will get less DR.someone told me that the extended DR starts at a bit higher ISO though,I dont have my K5 here while writing.I dont know this at all,so probably reason that having the iso start at 125 or a bit more in auto probably works for me is due to being more forgiving for user error-meaning wrong exposure when I cannot not predict exactly how the camera will work???http://www.dxomark.com/index.php/Cameras/Camera-Sensor-Database/Pentax/K5The reduction ought to be about 1 stop of DR lost for each stop of increased ISO, although this direct relationship doesn't always hold, but DRneverincreases as you increase ISO.In-camera JPG conversion throws away some of the raw DR, reducing it to about 9 stops. As the raw DR drops to about 9 stops at ISO1600, you'll get about 9 stops DR in JPG whatever ISO you set. -- ---GerryFirst camera 1953, first Pentax 1985, first DSLR 2006 http://www.pbase.com/gerrywinterbourne

henning kraggerudwrote:If you like me have quite frequent problems with underexposure in an rather unpredictable way on your K5 or other Pentax DSLRs(in my case especially with bright aperture lenses like older 1.4 lenses etc)I can relate to that.and dont want to keep changing exposure compentation when you changing from big apertures to small and a lot of other scenariosI've not experienced the lens or aperture dependency. It's more of a light and contrast dependency for me. I experience low exposure at low contrast, medium light level type of scenes, like daytime indoors.[ ... snipped high/low key adjustments ... ]I also use the high/low key adjustment, mostly to provide fill light in backlit situations where I can't use flash. It's basically a midtone adjustment as far as I can tell.Regards, --Anders.

Gerry Winterbournewrote:In-camera JPG conversion throws away some of the raw DR, reducing it to about 9 stops.That's true, for the default jpeg processing settings, i.e. contrast 0, high/low key 0, shadow and highlight protection off.However, if you turn everything to the max you can squeeze several extra stops of scene DR into your jpegs. I've been meaning to design and execute an experiment to measure the jpeg DR with the following settings:Custom image naturalcontrast -4, high/low key +4, shadow contrast +4, highlight contrast -4highlight protection onshadow protection 3I've used those kinds of settings to cope with high contrast scenes in-camera, like dark interiors with sunlit windows. I can then dial in -2 EC and make both the window and the faces and dark bits inside look OK at the same time, in camera. I even have a user setting called "max lift" for that.But that only works for me because I shoot for fun and can shoot jpeg at my own pace. If I was in a hurry I'd of course have to shoot raw and do all the "heavy lifting" (pun intended) in post.My point is, the in camera jpeg engine can be setup to do substantially more than 9 stops of DR.Regards, --Anders.

henning kraggerudwrote:Gerry Winterbournewrote:henning kraggerudwrote:when workin in TAv and and some other modes I also prefer to have auto iso dialed strating not from iso 100 but at least 125 or more,as that gives more dynamic range(or room for error),the way I shoot. Any body who have more experince with that? (Dynamic range at different ISOs)Raw DR always reduces as you increase ISO. Here's the DxOMark chart for the K-5.I assume that is not for extended iso, so from 80 to 100 you will get less DR.No. The best raw DR is at ISO80, and is about 1/2 stop wider than at ISO100.someone told me that the extended DR starts at a bit higher ISO though,I dont have my K5 here while writing.If you use the D-range highlight feature it increases ISO by 1 stop from whatever you have set ISO at. Thus, if you extend ISO down to 80 and select D-range highlights ISO goes automatically yo 160; if you have set ISO to 200 and select D-range ISO goes up to 400.If you are shooting JPG this setting doesn't actually increase DR but moves it slightly - it operates to preserve highlights. If you are shooting raw D-range reduces DR (because it increases ISO). It's usually better in raw to use -EV to protect highlights and open up yourself in PP.I dont know this at all, so probably reason that having the iso start at 125 or a bit more in auto probably works for me is due to being more forgiving for user error-meaning wrong exposure when I cannot not predict exactly how the camera will work???I don't think so. Remember that in TAv you choose both shutter speed and aperture so the camera looks at how much light it is receiving and calculates the ISO needed for correct exposure. The only thing Auto-ISO does in TAv is warn you if your chosen shutter speed and aperture can't get a correct exposure with the range of ISOs you've set.If your chosen TAv parameters would allow correct exposure at ISO80 then that's what the camera would use. If you've told the camera to go no lower than ISO125 it will tell you that it can't do what you want: if you still shoot at those parameters but ISO125 instead of ISO80 the shot will be overexposed so there's no protection on offer.If your chosen TAv parameters need ISO 800 then that's what the camera will use regardless of your low limit on Auto-ISO.http://www.dxomark.com/index.php/Cameras/Camera-Sensor-Database/Pentax/K5The reduction ought to be about 1 stop of DR lost for each stop of increased ISO, although this direct relationship doesn't always hold, but DRneverincreases as you increase ISO.In-camera JPG conversion throws away some of the raw DR, reducing it to about 9 stops. As the raw DR drops to about 9 stops at ISO1600, you'll get about 9 stops DR in JPG whatever ISO you set.

henning kraggerudwrote:Gerry Winterbournewrote:henning kraggerudwrote:when workin in TAv and and some other modes I also prefer to have auto iso dialed strating not from iso 100 but at least 125 or more,as that gives more dynamic range(or room for error),the way I shoot. Any body who have more experince with that? (Dynamic range at different ISOs)Raw DR always reduces as you increase ISO. Here's the DxOMark chart for the K-5.Thanks for the reply I assume that is not for extended iso,so from 80 to 100 you will get less DR.iso 80 DR = 14.12 is0 100 DR = 13.7 iso 125 DR = 13.5http://www.dxomark.com/index.php/Cameras/Camera-Sensor-Database/Pentax/K5#tabs-2 -- My PPG http://www.pentaxphotogallery.com/artists/andrewwaldram My Photo Stream http://www.flickr.com/photos/awaldram/ 1x.com http://1x.com/artist/awaldram/wall

asp1880wrote:Gerry Winterbournewrote:In-camera JPG conversion throws away some of the raw DR, reducing it to about 9 stops.That's true, for the default jpeg processing settings, i.e. contrast 0, high/low key 0, shadow and highlight protection off.However, if you turn everything to the max you can squeeze several extra stops of scene DR into your jpegs. I've been meaning to design and execute an experiment to measure the jpeg DR with the following settings:Custom image naturalcontrast -4, high/low key +4, shadow contrast +4, highlight contrast -4highlight protection onshadow protection 3I've used those kinds of settings to cope with high contrast scenes in-camera, like dark interiors with sunlit windows. I can then dial in -2 EC and make both the window and the faces and dark bits inside look OK at the same time, in camera. I even have a user setting called "max lift" for that.But that only works for me because I shoot for fun and can shoot jpeg at my own pace. If I was in a hurry I'd of course have to shoot raw and do all the "heavy lifting" (pun intended) in post.My point is, the in camera jpeg engine can be setup to do substantially more than 9 stops of DR.I'd like to see some evidence that this actually extends the DR rather than just altering the tonality within it. The DR page of the DPR review of the K-5 shows almost no variation of DR for the (relatively few) different settings it measured. http://www.dpreview.com/reviews/pentaxk5/14Note that the main effect of Highlight D-range is to move the DR rather than to widen it, although it does get fractionally wider.I'd be surprised if all these adjustments got the DR wider than about 10 stops - but happy if you can show that they do.

Gerry Winterbournewrote:henning kraggerudwrote:Gerry Winterbournewrote:henning kraggerudwrote:when workin in TAv and and some other modes I also prefer to have auto iso dialed strating not from iso 100 but at least 125 or more,as that gives more dynamic range(or room for error),the way I shoot. Any body who have more experince with that? (Dynamic range at different ISOs)Raw DR always reduces as you increase ISO. Here's the DxOMark chart for the K-5.I assume that is not for extended iso, so from 80 to 100 you will get less DR.No. The best raw DR is at ISO80, and is about 1/2 stop wider than at ISO100.someone told me that the extended DR starts at a bit higher ISO though,I dont have my K5 here while writing.If you use the D-range highlight feature it increases ISO by 1 stop from whatever you have set ISO at. Thus, if you extend ISO down to 80 and select D-range highlights ISO goes automatically yo 160; if you have set ISO to 200 and select D-range ISO goes up to 400.If you are shooting JPG this setting doesn't actually increase DR but moves it slightly - it operates to preserve highlights. If you are shooting raw D-range reduces DR (because it increases ISO). It's usually better in raw to use -EV to protect highlights and open up yourself in PP.I dont know this at all, so probably reason that having the iso start at 125 or a bit more in auto probably works for me is due to being more forgiving for user error-meaning wrong exposure when I cannot not predict exactly how the camera will work???I don't think so. Remember that in TAv you choose both shutter speed and aperture so the camera looks at how much light it is receiving and calculates the ISO needed for correct exposure. The only thing Auto-ISO does in TAv is warn you if your chosen shutter speed and aperture can't get a correct exposure with the range of ISOs you've set.Well you get what camera deems to be right exposureIf your chosen TAv parameters would allow correct exposure at ISO80 then that's what the camera would use. If you've told the camera to go no lower than ISO125 it will tell you that it can't do what you want: if you still shoot at those parameters but ISO125 instead of ISO80 the shot will be overexposed so there's no protection on offer.you are right,I didnt think it enough through, but in aperture or shutter mode it could make more sense I would think as camera could get what it deems to be right exposure both at 80 , 100 , 125 200 iso etc,and if you happen to disagree with camera you would have more headroom whit a higher starting point.If your chosen TAv parameters need ISO 800 then that's what the camera will use regardless of your low limit on Auto-ISO.http://www.dxomark.com/index.php/Cameras/Camera-Sensor-Database/Pentax/K5The reduction ought to be about 1 stop of DR lost for each stop of increased ISO, although this direct relationship doesn't always hold, but DRneverincreases as you increase ISO.In-camera JPG conversion throws away some of the raw DR, reducing it to about 9 stops. As the raw DR drops to about 9 stops at ISO1600, you'll get about 9 stops DR in JPG whatever ISO you set.

I think there is some confusion here between DR and 8bit Jepg.In theory a Jpeg can store infinite DR given a sparse enough bit spread from the tone curve.So yes adjusting tone curve can increase the DR fo a Jpreg image but it will always be at the expense of posterization.If you applied a standard Gamma 2.2 curve you would achieve something like 18 stops DR but the image would be unusable due to posterizationTherefore 9 stops is considered the correct balance between retained DR and Bit spread.I very much doubt any in camera setting will appreciably alter this as the results would be disappointing.If you want more DR your choices are RAW or hdr there is no free lunch from jpeg.Gerry Winterbournewrote:asp1880wrote:Gerry Winterbournewrote:In-camera JPG conversion throws away some of the raw DR, reducing it to about 9 stops.That's true, for the default jpeg processing settings, i.e. contrast 0, high/low key 0, shadow and highlight protection off.However, if you turn everything to the max you can squeeze several extra stops of scene DR into your jpegs. I've been meaning to design and execute an experiment to measure the jpeg DR with the following settings:Custom image naturalcontrast -4, high/low key +4, shadow contrast +4, highlight contrast -4highlight protection onshadow protection 3I've used those kinds of settings to cope with high contrast scenes in-camera, like dark interiors with sunlit windows. I can then dial in -2 EC and make both the window and the faces and dark bits inside look OK at the same time, in camera. I even have a user setting called "max lift" for that.But that only works for me because I shoot for fun and can shoot jpeg at my own pace. If I was in a hurry I'd of course have to shoot raw and do all the "heavy lifting" (pun intended) in post.My point is, the in camera jpeg engine can be setup to do substantially more than 9 stops of DR.I'd like to see some evidence that this actually extends the DR rather than just altering the tonality within it. The DR page of the DPR review of the K-5 shows almost no variation of DR for the (relatively few) different settings it measured. http://www.dpreview.com/reviews/pentaxk5/14Note that the main effect of Highlight D-range is to move the DR rather than to widen it, although it does get fractionally wider.I'd be surprised if all these adjustments got the DR wider than about 10 stops - but happy if you can show that they do.

Interesting discussion and I need to review it some more of it to learn a bit about the K5 because I am a new owner. Anyway, shooting with AEB, with two frames can be helpful is you aren't opposed to merging them if either doesn't deliver what you are looking for.As an HDR shooter(don't jump on me it isn't against the law) you can get great...and natural results with the K5. I have taken a few in the past couple of days and even though I am using under a very limited range +- the result is nothing short of good. In fact, I have no doubt it is better with its bracketing than the Canon 7D, which was up until now, my favorite DSLR for AEB shooting.

Gerry Winterbournewrote:asp1880wrote:[ ... snip ... ]My point is, the in camera jpeg engine can be setup to do substantially more than 9 stops of DR.I'd like to see some evidence that this actually extends the DR rather than just altering the tonality within it. The DR page of the DPR review of the K-5 shows almost no variation of DR for the (relatively few) different settings it measured. http://www.dpreview.com/reviews/pentaxk5/14Note that the main effect of Highlight D-range is to move the DR rather than to widen it, although it does get fractionally wider.I'd be surprised if all these adjustments got the DR wider than about 10 stops - but happy if you can show that they do.I wondered how to do this without a "Stouffer step wedge" that the dpreview lab has. Maybe generate the high contrasts by varying the shutter speed several multiple exposures? Then I gave up, fired up Pentax DCU 4 and loaded a darkish ISO 80 raw file.DCU has the same jpeg engine as the camera, right? With the same contrast and shadow compensation stuff as the camera. Here's what I did: a) Set DCU4 to custom image natural, all sliders centered. b) Locate a dark patch of the scene, where DCU4 renders (0,0,0).c) Set contrast to -4, high/low key +4, shadow compensation max ("max lift settings") d) Watch how the dark patch renders brighter now, like (20,20,20) e) Grab the exposure slider and reduce until the dark patch renders like in b)Well it won't render exactly like in b) because of noise, what awaldram refers to as posterisation. Even so, I find that I need to reduce exposure 1.5-2 stops to render the dark patch about the same as without the "jpeg max lift settings". But I'm eyeballing things late at night here and other people might call it differently.So I tentatively conclude that the "jpeg max lift settings" squeeze 1.5-2 stops additional DR into the jpeg?Is my method valid? Futzing in DCU4 instead of using a Stouffer wedge? Have I just misunderstood something important here?Eagerly awaiting clarification, --Anders.

asp1880wrote:Gerry Winterbournewrote:asp1880wrote:[ ... snip ... ]My point is, the in camera jpeg engine can be setup to do substantially more than 9 stops of DR.I'd like to see some evidence that this actually extends the DR rather than just altering the tonality within it. The DR page of the DPR review of the K-5 shows almost no variation of DR for the (relatively few) different settings it measured. http://www.dpreview.com/reviews/pentaxk5/14Note that the main effect of Highlight D-range is to move the DR rather than to widen it, although it does get fractionally wider.I'd be surprised if all these adjustments got the DR wider than about 10 stops - but happy if you can show that they do.I wondered how to do this without a "Stouffer step wedge" that the dpreview lab has. Maybe generate the high contrasts by varying the shutter speed several multiple exposures? Then I gave up, fired up Pentax DCU 4 and loaded a darkish ISO 80 raw file.DCU has the same jpeg engine as the camera, right? With the same contrast and shadow compensation stuff as the camera. Here's what I did: a) Set DCU4 to custom image natural, all sliders centered. b) Locate a dark patch of the scene, where DCU4 renders (0,0,0).c) Set contrast to -4, high/low key +4, shadow compensation max ("max lift settings") d) Watch how the dark patch renders brighter now, like (20,20,20) e) Grab the exposure slider and reduce until the dark patch renders like in b)Well it won't render exactly like in b) because of noise, what awaldram refers to as posterisation. Even so, I find that I need to reduce exposure 1.5-2 stops to render the dark patch about the same as without the "jpeg max lift settings". But I'm eyeballing things late at night here and other people might call it differently.So I tentatively conclude that the "jpeg max lift settings" squeeze 1.5-2 stops additional DR into the jpeg?Is my method valid? Futzing in DCU4 instead of using a Stouffer wedge?The method itself might be valid in principle; but what you've reported doesn't actually support your conclusion.Have I just misunderstood something important here?What you've missed is that DR is arangefrom darkest to brightest and you've told us nothing about the bright end of your test shot - except that as you say the shot was "darkish" we can assume that the bright end may not have been near clipping.For your method to be valid the first prerequisite is that the scene itself had a wide DR - ideally the 14 stops that a raw file can capture or wider. That way we'll know that there is plenty of DR for the JPG conversion to work on.Next the scene must be exposed so that the in-camera JPGjustreaches the RH end of the histogram. A reading of 255, 255, 255 might be slightly clipped so we'd want something like 253/4, ... just to be sure we know where the top end of the DR is. You'd have to shoot RAW+ and compare the histograms to confirm that PDCU does use the same JPG engine; and, of course, the camera settings should be Natural and zeroed throughout.After that follow your procedure but with this modification: bring the highlight end to the same 253/4, ... that you started with. This ensures that you haven't moved the top end of the DR.Now you can examine the reading for the dark spot you selected to see what has happened to the bottom end of the DR.